[R-meta] [netmeta] Error message in network meta-analysis: "Problem with multi-arm studies: Studies with inconsistent treatment estimates"

[Gerta Ruecker]

Dear Rebecca,

First to the theoretical background of this error message: If you have a 
three-arm study, treatment effects within the study must be consistent. 
This means that for three arms A, B, C you have mean values x_A, x_B, 
x_C of your score. The contrasts (differences) then are

  * for A vs B: d_AB = x_A - x_B
  * for A vs C: d_AC = x_A - x_C
  * for B vs C: d_BC = x_B - x_C

That is, you have relations between the three contrasts, e.g., d_AC = 
d_AB + d_BC. If the data do not fulfill these relations, an error 
message is given.

However, I have a problem with understanding your data structure, and 
thus I don't fully understand how this can happen in your case. You say 
you have three-arm studies, but how are they entered? Unfortunately, the 
first three studies in your extract of the data are all two-arm. How do 
the three-arm studies look in the data?

It would be easier to help if we could see the whole data set or at 
least data for a three-arm study.

Best,

Gerta


Am 07.02.2020 um 16:42 schrieb Rogasch, Rebecca Manuela Madeleine:
> Dear experts,
>
> I'm seeking advice regarding an error message in network meta-analysis using the netmeta package.
>
> To briefly outline the objective and describe the data structure:
> I am conducting a network meta-analysis on the effectiveness of PTSD intervention programs.
> The primary outcome is the PTSD total symptom score (=continuous outcome).
> All studies are randomized-controlled trials. There are 8 three-arm studies in my data set and 64 two-arm studies.
> I calculated the SMD (post-post) from means and SDs.
>
> Here is an example for the data structure:
>
>
>   studlab sampleID  OutcID   TR1   TR2    Mt2TR1  SDt2TR1  Mt2TR2  SDt2TR2  n_TR1 n_TR2 n_total    smd    se
>
>         1      101 10101     EMDR   WL    22.87    20.27    54.21   16.26    15    14      29    -1.65  0.19
>
>       100    10001 1000101   NET    WL    55.07    27.01    76.86   17.14    19    19      38    -0.94  0.12
>
>       104    10401 1040101   NET   PsEd   19.10    11.70    21.20    9.40    17    12      29    -0.19  0.14
>
>
>
> When applying the netmeta function I get the following error message:
>
>> mn1 <- netmeta(smd, se, TR1, TR2, studlab, data=dat3, sm="SMD", comb.random=TRUE)
> Error: Problems in multi-arm studies!
> - Studies with inconsistent treatment estimates: '107', '121', '130', '15', '22', '34', '36', '87'
> - Please check original data used as input to netmeta().
> - For more details, re-run netmeta() with argument details.chkmultiarm = TRUE.
> - Argument 'tol.multiarm' in netmeta() can be used to relax consistency
> assumption for multi-arm studies (if appropriate).
>
> My questions are as follows:
>
>    1.  I am not able to make sense of the error message regarding inconsistency within the 3-arm-studies. Can anyone help?
>    2.  When setting the argument tol.multiarm = 0.05 it runs without any error message.
>
> However, I am not sure what the argument does, exactly, and haven't been able to find any detailed information.
>
> Does anyone know what the argument does or where I can find further information?
>
> Looking forward to hearing from you!
> Best regards,
> Rebecca
>
> 	[[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-meta-analysis mailing list
> R-sig-meta-analysis using r-project.org
> https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis

-- 

Dr. rer. nat. Gerta Rücker, Dipl.-Math.

Institute of Medical Biometry and Statistics,
Faculty of Medicine and Medical Center - University of Freiburg

Stefan-Meier-Str. 26, D-79104 Freiburg, Germany

Phone:    +49/761/203-6673
Fax:      +49/761/203-6680
Mail:     ruecker using imbi.uni-freiburg.de
Homepage: https://www.imbi.uni-freiburg.de/persons/ruecker/person_view


	[[alternative HTML version deleted]]